Headlong Into Heavy Reading

Over the past three years, Light Reading has issued a monthly series of Insiders, for those of our readers willing to put up a few dollars for in-depth analysis of next-generation telecom topics.

Hungry for information, our customers have welcomed our efforts and come back for more. And so we have turned our attention to bigger things: not merely longer Light Reading reports, but a whole new domain for exhaustive, yet readable, heavyduty analysis that can sustain real-world business planning and operations. With that in mind, we bring you Heavy Reading.

Heavy Reading’s goal is to furnish our readers with the comprehensive information and analysis of products (and markets) that will enable them to make the right decisions about their network infrastructures – in terms of both technology and business.

Now, there’s a couple of common ways to go about the business of market research – gathering statistics or taking industry soundings, and both tend to be rather thankless.

On the one hand, you can collect industry data to build complex, biz-school-caliber forecasting models. These can help managers make the case for building that new plant or investing to capture new market share – or they may simply illustrate how much has been surrendered to competitors and suggest a way of clawing it back. That kind of research, though perhaps a just a wee bit dull (a volume whose narrative climax is the description of 9.5% CAGR will never be accused of being a page-turner), has its uses – and it does give the customers great charts for presentations to their boards or VCs.

Sounding the industry is another thing altogether: You listen a great deal to the right people in the industry until you are ready to start talking back to it in a useful way, with reasonable acumen and authority. In other words, you cast your net and go trawling – in the hopes that you’ll be able to scavenge what others are missing, or at least aren’t getting quite right, due to inadequate information. Then you set about assembling your facts. Alas, a well honed report can turn out to be a bit of a challenge for many readers, who, when lifting a 4-pound hardback, would much prefer to enter a world of battling wizards or Hollywood gossip than that of OEO switching.

Now, there’s a third way: Heavy Reading undertakes market research that provides real numbers, vital facts, and practical analysis – and manages to get to the point before page 100; surveys that combine detailed and expansive coverage with the industry expertise of senior, heavyweight analysts.

The Heavy Reading team includes Geoff Bennett, the man behind Light Reading University and our newly minted Chief Technologist; and Graham Beniston, one of our senior Analysts at Large. I’ve assumed the title of Chief Analyst and relinquished the title of Research Director at Light Reading.

The split with Light Reading is important and not just a matter of a title transfer. We’re creating a separate entity here, one that can leverage the reach of Light Reading when conducting surveys and reaching out to the industry, but allows no leakage of the confidential information vendors and service providers may share with our staff.

We’ve put together a first-rate lineup of reports for the coming year. In our first report – Multiservice Provisioning Platforms: Empowering the Metro Edge – Analyst at Large Tim Hills has accomplished what no marketing manager could ever hope to on his or her own: fill in every cell on a table with 75(!) columns and 45 rows that amasses the details on every MSPP vendor in the world. This report isn’t so much a study of metro market development (we know it’s doing well enough, given the circumstances) as a scientific dissection of the ill-defined MSPP.

For every system developer, chip designer, software architect, service provider, and systems integrator, this report should serve a dual role of market analysis and desktop reference. Want to know what percentage of vendors support RPR, LCAS, or 10-Gigabit Ethernet? It’s in there. And, amazingly, it’s pretty easy to find. The spreadsheet is an unwieldy beast, but, thanks to HTML, we’ve been able to create a very dynamic Web page that users can easily navigate. Furthermore, users will be able to customize their own tables, based on specific requests and requirements. This is how market research is supposed to work.

What’s unique about this report is the completeness of the survey – with every possible feature documented – and the process of tying those results to the larger issues overhanging the market today, e.g., identifying when Ethernet will become the predominant service interface in the metro.

As with any research project that includes filling in over 6,000 data fields in a spreadsheet, interesting themes appear once you’ve nailed down the data. This report, supports the following key findings (among others):

  • Transport-based MSPPs must support GFP/G.7041 and LCAS in order to effectively support packet services over Sonet/SDH.

  • Growth in MSPP sales has given rise to an important new product category – MSPP core aggregation platforms.

  • Both HVAC capacity and safety regulations place practical limits on products’ theoretical port counts/capacity.

  • MSPPs featuring schemes like BLSR and ULSR allow redundant circuits to be used for live data, doubling the capacity of metro networks.

  • Prices vary from $5,000 for a bare-bones MSPP installed in a metro access network to over $400,000 for a fully outfitted core MSPP.

  • MPLS is gaining support from MSPP vendors as a key mechanism for enabling packet services, QOS, and traffic engineering in the metro.

  • There is only limited support for SAN protocols – which is surprising, in view of the growing interest in enterprise data security and recovery.
The report doesn’t stop at the spreadsheet. It can’t, really, because the MSPP space is an odd one and needs a lot of help in self-definition. There are many ways to slice and dice it (perhaps too many) but the effort is important.

The MSPP, in its initial incarnation, was a Multiservice Sonet/SDH Provisioning Platform, which lent it a certain circuit-centric air. Then, thanks to new chip technology that supports packetization and emulation of nearly every possible service, pure-packet solutions evolved. This led to a split of MSPPs into transport-based or data-based. As a way to classify systems based on their internal architecture, MSPPs are broadly defined in our report as either one or the other.

We identify whether the MSPPs are based on circuit switches, packet switches, or some kind of hybrid; and we ask whether they support true TDM, with circuit grooming, or if that is all emulated. This classification process is tough, because many vendors evolve their products to suit the requirements of their lead customers. Some MSPPs started out looking a lot like data-based solutions, but lately there’s enough circuit-switching going on under the hood to move them into the transport-based category (even if wedged in there a bit uncomfortably).

When you throw in WDM, the classification process becomes even more complicated. Metro WDM has arguably undergone three evolutionary periods: the first being the stripped-down long-haul DWDM solution; the second, the metro-optimized ring solution; and the third, upon us now, the multiservice DWDM platform, with optical layer flexibility. The problem is that the third-gen product blurs the boundary between metro DWDM and the MSPP.

Metro WDM systems perform sub-lambda aggregation, switching, and grooming, so they have much of the same functionality as the MSPP. However, they differ in the convergence layer, which for WDM, is the lambda rather than the OC48 or OC192 or 10-Gig Ethernet. Their purpose, and the value they bring to certain carrier requirements, is to support a mix of full-rate protocol transport (such as Fibre Channel) and sub-rate services (Ethernet, DS-3, OC-n) from a common shelf on a single fiber. Add to this the ability to add, drop, and express wavelengths at any node; hitlessly expand capacity; and provide protection both of services and of the network in the optical layer (thanks to an integrated ROADM) and you then have a system that stretches the boundaries of the MSPP to the point of pain.

Who’s in this category? Well, I’d say companies such as PacketLight Networks, Photuris Inc., Tropic Networks Inc., and perhaps Internet Photonics Inc. Some of Ciena Corp.’s gear seems close. WDM is not an option in these products, but fundamental to their architectures, at the layer where they interface to the transport network. Other large vendors will likely have products that also fit this bill, as will suppliers of MSPPs, third-gen WDM, and anything else that sells into this finicky market. NFOEC will reveal more players, I’m sure.

That’s a taste of the MSPP report. So what’s coming next from Heavy Reading? Lots!

We have a report coming from Geoff Bennett on Incumbent Vendor Strategies; another report from Graham Beniston on Broadband Infrastructures for DSL Deployment; and I’m working on The Future of Sonet/SDH – wherein I’ll attempt to answer the question of when Sonet/SDH will finally shuffle off its mortal coil, by looking at the viability and vitality of services carried over those networks. The short answer today is “not anytime soon,” but I’ll be putting much more meat on the bone, describing service definitions for Ethernet over transport networks, storage over transport networks, and the role of GFP, VCAT, LCAS, and GMPLS – the four pillars supporting the recently scrubbed façade of next-gen Sonet/SDH.

Stay tuned!

— Scott Clavenna, Chief Analyst, Heavy Reading

Be the first to post a comment regarding this story.
Sign In